Fault detection in cyber-physical systems
a cyber-physical system and fault detection technology, applied in the field of fault detection in cyber-physical systems, can solve the problems of substantial safety problems, repair costs, complex systems of vehicles,
Pending Publication Date: 2021-11-11
NEC LAB AMERICA
View PDF0 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Benefits of technology
This patent describes a method and system for training a neural network model to detect and identify faults in a vehicle's normal state and to generate a score that indicates how similar or dissimilar the input data is to the fault state training data. This system can be used in vehicles to help prevent accidents and improve safety.
Problems solved by technology
Vehicles are complex systems and including a variety of different functional components, such as the engine, battery, transmission, etc.
Faults can lead to substantial safety problems and repair costs.
Method used
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment Construction
[0016]Modern vehicles are equipped with a variety of electronic control units (ECUs), each of which may control a small operational unit of the vehicle, and may report the state of the operational unit. Faults in the vehicle's systems may be predicted and prevented based on data from the ECUs, which can prevent damage to the vehicle and loss of life. Using time series information generated by the ECUs, faults in the vehicle may be predicted for a given time period, with the fault being labeled as to a likely vehicle sub-system that is responsible.
[0017]This may be performed using a trained neural network model, using a first set of training data, Sn, which represents time series information from ECUs under normal operating conditions, and Sf, which represents time series information from the ECUs under a fault operating condition. The trained neural network model may be used to provide labels for new time series information Si from the ECUs of a car i, during a time period from t1 t...
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More PUM
Login to View More Abstract
Methods and systems for training a neural network model include processing a set of normal state training data and a set of fault state training data to generate respective normal state inputs and fault state inputs that each include data features and sensor correlation graph information. A neural network model is trained, using the normal state inputs and the fault state inputs, to generate a fault score that provides a similarity of an input to the fault state training data and an anomaly score that provides a dissimilarity of the input to the normal state training data.
Description
RELATED APPLICATION INFORMATION[0001]This application claims priority to U.S. Patent Application No. 63 / 021,291, filed on May 7, 2020, incorporated herein by reference in its entirety. This application is related to an application entitled “DEEP LEARNING OF FAULT DETECTION IN ONBOARD AUTOMOBILE SYSTEMS”, having attorney docket number 21005, and which is incorporated by reference herein in its entirety.BACKGROUNDTechnical Field[0002]The present invention relates to fault detection in cyber-physical systems, and, more particularly, to the use sensor data from electronic control units on a vehicle to detect faults in the vehicle.Description of the Related Art[0003]Vehicles are complex systems and including a variety of different functional components, such as the engine, battery, transmission, etc. Faults can lead to substantial safety problems and repair costs.SUMMARY[0004]A method for training a neural network model includes processing a set of normal state training data and a set of...
Claims
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More Application Information
Patent Timeline
Login to View More IPC IPC(8): G06N3/08G06N3/04G07C5/08
CPCG06N3/08G07C5/0808G06N3/0445G07C5/008B60W50/0205G06N3/084G06N3/044G06N3/045G07C5/085G06N3/088B60W2710/18B60W50/035B60W50/038B60W2710/06
Inventor TANG, LUANCHEN, HAIFENGCHENG, WEIRHEE, JUNGHWANKAMIMURA, JUMPEI
Owner NEC LAB AMERICA